OCR API

The OCR (Optical Character Recognition) API extracts text from images and documents with high accuracy and supports structured data extraction for specific document types.

Base URL

https://api.zaits.net/v1/ocr

Authentication

All requests require your API key in the Authorization header:

Authorization: Bearer YOUR_API_KEY

Text Extraction

Extract text from any document or image.

Endpoint

POST /v1/ocr/extract

Parameters

Parameter

Type

Required

Description

image

file

✅

Document image (JPG, PNG, PDF)

language

string

❌

Language code (e.g., "en", "es", "auto")

output_format

string

❌

Output format: text, json, structured

Supported Languages

auto - Auto-detect (default)
en - English
es - Spanish
fr - French
de - German
it - Italian
pt - Portuguese
And 100+ more languages

Request Example

curl -X POST https://api.zaits.net/v1/ocr/extract \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -F "[email protected]" \
  -F "language=en" \
  -F "output_format=structured"

const formData = new FormData();
formData.append('image', file);
formData.append('language', 'en');
formData.append('output_format', 'structured');

const response = await fetch('https://api.zaits.net/v1/ocr/extract', {
  method: 'POST',
  headers: {
    'Authorization': 'Bearer YOUR_API_KEY'
  },
  body: formData
});

const result = await response.json();

import requests

files = {'image': open('document.jpg', 'rb')}
data = {
    'language': 'en',
    'output_format': 'structured'
}

response = requests.post(
    'https://api.zaits.net/v1/ocr/extract',
    headers={'Authorization': 'Bearer YOUR_API_KEY'},
    files=files,
    data=data
)

result = response.json()

Response

{
  "success": true,
  "data": {
    "text": "Complete extracted text content from the document...",
    "confidence": 0.95,
    "language_detected": "en",
    "page_count": 1,
    "processing_time": 2.1,
    "regions": [
      {
        "text": "Header Text",
        "confidence": 0.98,
        "bounding_box": {
          "x": 10,
          "y": 20,
          "w": 200,
          "h": 30
        }
      },
      {
        "text": "Body paragraph content...",
        "confidence": 0.94,
        "bounding_box": {
          "x": 10,
          "y": 60,
          "w": 400,
          "h": 120
        }
      }
    ],
    "metadata": {
      "image_dimensions": {
        "width": 1200,
        "height": 800
      },
      "dpi": 300
    }
  }
}

Document-Specific OCR

Extract structured data from specific document types with intelligent field recognition.

Endpoint

POST /v1/ocr/extract/document

Parameters

Parameter

Type

Required

Description

image

file

✅

Document image

document_type

string

✅

Document type (see supported types below)

extract_tables

boolean

❌

Extract table data (default: false)

Supported Document Types

invoice - Business invoices
receipt - Purchase receipts
contract - Legal contracts
form - Government forms
bank_statement - Bank statements
tax_document - Tax forms
medical_report - Medical documents
insurance_claim - Insurance forms

Request Example

curl -X POST https://api.zaits.net/v1/ocr/extract/document \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -F "[email protected]" \
  -F "document_type=invoice" \
  -F "extract_tables=true"

Response (Invoice Example)

{
  "success": true,
  "data": {
    "document_type": "invoice",
    "confidence": 0.92,
    "fields": {
      "invoice_number": "INV-2025-001",
      "date": "2024-01-15",
      "due_date": "2024-02-14",
      "total_amount": 1234.56,
      "currency": "USD",
      "tax_amount": 123.45,
      "subtotal": 1111.11,
      "vendor": {
        "name": "Acme Corporation",
        "address": "123 Business St, City, State 12345",
        "tax_id": "123456789"
      },
      "customer": {
        "name": "John Doe",
        "address": "456 Customer Ave, City, State 67890"
      },
      "line_items": [
        {
          "description": "Product A",
          "quantity": 2,
          "unit_price": 500.00,
          "total": 1000.00
        },
        {
          "description": "Service B",
          "quantity": 1,
          "unit_price": 111.11,
          "total": 111.11
        }
      ]
    },
    "tables": [
      {
        "rows": 3,
        "columns": 4,
        "data": [
          ["Item", "Qty", "Price", "Total"],
          ["Product A", "2", "$500.00", "$1000.00"],
          ["Service B", "1", "$111.11", "$111.11"]
        ]
      }
    ],
    "raw_text": "Full extracted text content...",
    "processing_time": 3.2
  }
}

Receipt OCR

Specialized OCR for retail receipts with merchant and item recognition.

Endpoint

POST /v1/ocr/extract/receipt

Parameters

Parameter

Type

Required

Description

image

file

✅

Receipt image

currency

string

❌

Expected currency (e.g., "USD", "EUR")

extract_items

boolean

❌

Extract individual items (default: true)

Request Example

curl -X POST https://api.zaits.net/v1/ocr/extract/receipt \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -F "[email protected]" \
  -F "currency=USD" \
  -F "extract_items=true"

Response

{
  "success": true,
  "data": {
    "merchant": {
      "name": "SuperMart",
      "address": "789 Shopping Blvd, City, State",
      "phone": "(555) 123-4567"
    },
    "transaction": {
      "date": "2024-01-15",
      "time": "14:32:00",
      "transaction_id": "TXN123456"
    },
    "totals": {
      "subtotal": 42.00,
      "tax": 3.67,
      "total": 45.67,
      "currency": "USD"
    },
    "payment": {
      "method": "Credit Card",
      "card_last_four": "1234",
      "amount_paid": 45.67,
      "change": 0.00
    },
    "items": [
      {
        "description": "Apples - Organic",
        "quantity": 2,
        "unit_price": 3.99,
        "total": 7.98
      },
      {
        "description": "Bread - Whole Wheat",
        "quantity": 1,
        "unit_price": 4.50,
        "total": 4.50
      },
      {
        "description": "Milk - 2% Gallon",
        "quantity": 1,
        "unit_price": 4.29,
        "total": 4.29
      }
    ],
    "confidence": 0.91,
    "processing_time": 1.8
  }
}

ID Card OCR

Extract information from government-issued ID cards and driver's licenses.

Endpoint

POST /v1/ocr/extract/id

Parameters

Parameter

Type

Required

Description

image

file

✅

ID card image

document_type

string

❌

ID type: drivers_license, passport, national_id

country

string

❌

Country code (e.g., "US", "CA", "MX")

Request Example

curl -X POST https://api.zaits.net/v1/ocr/extract/id \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -F "image=@drivers_license.jpg" \
  -F "document_type=drivers_license" \
  -F "country=US"

Response

{
  "success": true,
  "data": {
    "document_type": "drivers_license",
    "country": "US",
    "state": "CA",
    "personal_info": {
      "first_name": "John",
      "middle_name": "Michael",
      "last_name": "Doe",
      "full_name": "John Michael Doe",
      "date_of_birth": "1990-01-01",
      "gender": "M",
      "height": "6'0\"",
      "weight": "180 lbs",
      "eye_color": "BRN",
      "hair_color": "BLK"
    },
    "address": {
      "street": "123 Main Street",
      "city": "Los Angeles",
      "state": "CA",
      "zip_code": "90210"
    },
    "document_info": {
      "id_number": "D1234567",
      "issue_date": "2020-01-15",
      "expiry_date": "2028-01-15",
      "class": "C"
    },
    "confidence": 0.94,
    "processing_time": 2.3
  }
}

Document Authenticity Verification

Verify document authenticity using advanced security feature detection.

Endpoint

POST /v1/ocr/verify/authenticity

Parameters

Parameter

Type

Required

Description

image

file

✅

Document image

document_type

string

✅

Type of document to verify

country

string

❌

Country of origin

Supported Document Types for Verification

passport - Passports
drivers_license - Driver's licenses
national_id - National ID cards
birth_certificate - Birth certificates
social_security_card - SSN cards

Request Example

curl -X POST https://api.zaits.net/v1/ocr/verify/authenticity \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -F "[email protected]" \
  -F "document_type=passport" \
  -F "country=US"

Response

{
  "success": true,
  "data": {
    "is_authentic": true,
    "confidence": 0.92,
    "risk_score": 0.08,
    "security_features": {
      "watermark_detected": true,
      "micro_printing": true,
      "holographic_elements": true,
      "uv_features": false,
      "rfid_chip": true,
      "barcode_validation": true
    },
    "authenticity_checks": [
      {
        "check": "Font Analysis",
        "status": "pass",
        "confidence": 0.95
      },
      {
        "check": "Layout Verification",
        "status": "pass", 
        "confidence": 0.89
      },
      {
        "check": "Security Pattern",
        "status": "pass",
        "confidence": 0.92
      }
    ],
    "anomalies": [],
    "processing_time": 4.1
  }
}

Error Responses

Common OCR Errors

Error Code

HTTP Status

Description

no_text_detected

400

No readable text found in the image

poor_image_quality

400

Image quality too low for OCR

unsupported_language

400

Specified language not supported

document_type_mismatch

400

Document doesn't match specified type

image_orientation_error

400

Image needs to be rotated

multi_page_not_supported

400

Multi-page documents not supported for this endpoint

Error Response Format

{
  "success": false,
  "error": {
    "code": "poor_image_quality",
    "message": "The image quality is too low for reliable OCR processing",
    "details": {
      "image_resolution": "150x200",
      "min_recommended": "300x400",
      "suggestions": [
        "Use higher resolution images (min 300 DPI)",
        "Ensure good lighting and contrast",
        "Avoid blurry or distorted images"
      ]
    }
  }
}

Best Practices

Image Requirements

Format: JPG, PNG, PDF (single page)
Size: Max 10MB per file
Resolution: Min 300 DPI recommended
Orientation: Upright, properly aligned
Lighting: Even, well-lit images
Focus: Sharp, clear text

Performance Tips

Crop to relevant area to improve accuracy
Use high contrast images (dark text on light background)
Rotate images to correct orientation before uploading
Compress large files while maintaining quality
Use appropriate document type for structured extraction

Language Detection Tips

Set language=auto for automatic detection
Specify language when known for better accuracy
Multi-language documents may have lower confidence
Some languages require specific fonts for optimal results

Next: Document Signing API

PreviousFace Matching API NextDocument Signing API

Last updated 2 months ago